cancel
Showing results for 
Search instead for 
Did you mean: 

STM32F103C8 PWM and clock slow down and MCU restarts in noisy environment.

Sujoy_B
Associate II

Hi, 
We are using an STM32F103 as a controller for a Half-bridge dc-dc convertor running at 110Khz. We are facing an issue where the microcontroller is resetting every minute when output crosses 12A. 

We disconnected all peripherals, removed all protections and kept only the PWM and reset problem sustains itself. On further inspection, we found that the PWM duration increases for a few cycles and then the microcontroller resets. It seems to us that the microcontroller clock slows down and then finally stops forcing a reset. We do not have watchdog timer implemented.
We are using a 8MHz crystal. 

Please check attached image (CH1 - reset line, CH2 - Independent PWM, CH3 - Half bridge high side) 

The fun fact is that if we place the microcontroller on an external PCB and connect to the gate driver with wires, we do not see the microcontroller resetting. (even without the crystal). 

Sometimes, attaching an oscilloscope probe (1Mohm) also prevents restart. 

Currently, the microcontroller is grounded and the PCB is two layered. We have temporarily removed power ground from the ground plane and connected a wire to act as a return path.  

The PCB is two layered, snapshot attached. 
Any advice will be appreciated and please feel free to ask for more data. 

Regards,
Sujoy.

1 ACCEPTED SOLUTION

Accepted Solutions
mƎALLEm
ST Employee

Hello,

It looks like EMI issue / "ground bounce" phenomenon due to the high current switching (12A) creating an induced voltage v = Ldi/dt in the ground plane.

You have to adopt a "Star ground" configuration. The power ground and signal ground should meet at exactly one point to break any common ground impedance path. 

To give better visibility on the answered topics, please click on "Accept as Solution" on the reply which solved your issue or answered your question.

View solution in original post

1 REPLY 1
mƎALLEm
ST Employee

Hello,

It looks like EMI issue / "ground bounce" phenomenon due to the high current switching (12A) creating an induced voltage v = Ldi/dt in the ground plane.

You have to adopt a "Star ground" configuration. The power ground and signal ground should meet at exactly one point to break any common ground impedance path. 

To give better visibility on the answered topics, please click on "Accept as Solution" on the reply which solved your issue or answered your question.